Sabitlenmiş Tweet
Crazyox
60.9K posts
The key line from the Claude Code creator's demo:
A lot of his code these days is written by routines.
His work shifted from prompting Claude Code to writing routines that do the prompting for him.
The routine watches GitHub issues, starts a session, verifies its work in the browser, auto-fixes CI, and shepherds the PR all the way to merge.
That's what an agent looks like once it enters a real workflow.
That workflow lives on 12 layers. I mapped them here:
Vox@Voxyz_ai
English
>2026 has too many AI agent tools to track
>after saving too many launches, i realized i was comparing completely different things
>ChatGPT is the brain, MCP is the port, memory is the notebook, eval is the health check, control plane is the keycard
>so i split agents into 12 layers, and asked one question for each: where does real work break when this is missing?
>a few days with the map, and new launches start looking less like noise and more like coordinates
Vox@Voxyz_ai
English
Music video generation is one of the fastest growing and most exciting spaces.
@GeminiApp launched a series of new products: Gemini Flash, Omni, AI studio, Managed Agents.
Today, at the Google I/O hackathon, I got to build with them all!
Omnidesk is a music video workstation to turn your best ideas across all modals to the music video of choice.
Code is open sourced (MIT), and demo video is available for you take a look through the possibilities.
What is your vibe? How’s your experience with Google’s releases?
English
Greg Isenberg broke down a 35-step playbook for building AI startups from zero to $1M+/year. 4 businesses, all profitable, no VC money
I compressed it into 12 rules that actually matter:
1. ride a trend, don't create one: find a niche where demand already exists but tools are 5 years behind. if people are complaining in reddit threads and facebook groups.. that's your market
2. audience before product: master ONE platform. set notifications for 10 niche leaders, reply with real insight (not "great post!"), gain 5-10 followers daily. this compounds into an unfair advantage when you launch
3. validate with wallets, not surveys: if nobody will pre-pay at 50-70% off for something that doesn't exist yet.. they won't pay full price either. your audience becomes your investors
4. vibe code the MVP: use Cursor/Bolt/V0 to ship v1 in days not months. the goal isn't perfect code, it's proving people will use it. polish comes after revenue
5. keep the team at zero: AI is your co-founder now. Manus for research, ChatGPT for PRDs, Claude Design for design, Cursor for code. most $1M/yr businesses in 2026 run with 1-3 people max
6. automate before you hire: find the 3-step process you do 10x a day that takes 5 minutes each time. automate that first. Lindy, Gumloop, Zapier. work toward 90%+ automation
7. retention before growth: fixing churn is 10x more valuable than doubling acquisition. run retention sprints before growth sprints. most founders do this backwards and wonder why revenue flatlines
8. modular pricing kills one-size-fits-all: free tier for top of funnel, $29 for individuals, $299 for teams, $3K for enterprise. let the product sell itself at every level
9. partner with creators instead of buying ads: offer 1-20% equity or 20-50% rev share to creators with your audience. one partnership can outperform 6 months of paid marketing
10. build free tools for distribution: a public-facing calculator, checker, or generator drives organic traffic AND trains AI search to recommend you. this is the new SEO
11. think portfolio, not single product: once business #1 is profitable, repeat the process. share infrastructure, cross-promote, create a flywheel. Walt Disney didn't build one ride
12. ship something new every 30 days: a culture of shipping beats a culture of planning. new MVP monthly, acquire underperforming products with distribution upside, recruit operators to run them
what actually compounds in 2026:
- audience before product, not the other way
- pre-selling before building
- 1-person teams running $1M businesses with AI
- retention over acquisition, always
- portfolios over single bets
- shipping over strategizing
the cost of building has never been lower.. billions of people with credit cards are reachable through social media
the only bottleneck left is you actually starting
full 30-min breakdown from @gregisenberg attached below ↓
study this
P.S. left 20 not taken startup ideas below which you can take and start growing
Ronin@DeRonin_
The 20 BIGGEST startup ideas I'd build if I had 20 lives 1. biggest subscriptions: kill agent. average person bleeds $219/month on charges they forgot about. connect bank, show usage, cancel in one click. you keep 20% of what you save them 2. biggest insurance: AI that fights denied claims for you. insurers reject 30-40% on first try on purpose hoping you give up. most people do 3. biggest death tech: digital afterlife manager. you have 200+ accounts. crypto, subscriptions, passwords. when you die, half of it just.. disappears. 48% of americans have zero plan for this 4. biggest negotiation: AI that handles price negotiations over email. car deals, medical bills, salary offers. most people leave $5k-$50k/year on the table because they hate the back and forth 5. biggest fintech: AI tax strategist. not filing, strategy. what your $2k/hour CPA knows but packaged for $99/month 6. biggest debt: AI settlement negotiator. americans owe $1.1T in credit card debt. collection agencies buy your debt for pennies on the dollar and then charge you full price. AI that negotiates settlements at 30-50 cents on the dollar 7. biggest manufacturing: micro-factory OS. 3D printers went from $100k to $200 but the software stayed at $50k. thousands of people making real products from garages with spreadsheets 8. biggest fraud prevention: B2B payment verification. 76% of companies got hit last year. $133k average per incident. AI that checks every outgoing payment before it leaves 9. biggest construction: permit automation. you're waiting 3-6 months for paperwork that AI can fill in minutes. 1.5M permits/year in the US 10. biggest government: AI benefits navigator. $140B in federal benefits go unclaimed every year. people qualify but can't get through the paperwork. 47 pages to apply for programs that exist to help you 11. biggest local biz: AI reputation manager. one 1-star google review can kill a small business overnight. monitor everything, auto-respond, push positive results up. most owners have no idea what's being said about them 12. biggest freelance: AI contract reviewer. 70M+ freelancers signing stuff they don't fully read. flag risks in 30 seconds, charge $19/month 13. biggest field work: voice CRM. electricians and plumbers don't sit at desks. they need a CRM they can talk to while driving between jobs. 60M+ workers, nothing good exists 14. biggest proptech: phone-based property inspector. point at a room, get a full report. the $5B inspection industry still literally uses clipboards 15. biggest divorce: separation logistics AI. 750k divorces per year in the US. asset splitting, custody scheduling, document filing. lawyers charge $15-30k for what's mostly paperwork and coordination 16. biggest renovation: AI cost estimator. homeowners always get 3-5 completely different quotes from contractors. scan your rooms, know the real number before calling anyone 17. biggest compliance: EU e-invoicing tools. 10M+ SMBs are forced to go digital by 2027 and have no clue how. $22B market btw 18. biggest immigration: visa application automation. people spend $5-15k on immigration lawyers for paperwork that follows a process. millions of applications per year, all manually filled 19. biggest legal: AI dispute resolver. small claims under $10k take months in court and cost more in lawyer fees than the claim itself. online AI mediation that settles it in days for a flat fee 20. biggest healthcare: AI second opinion. 12M americans get misdiagnosed every year. 795,000 die or get permanently disabled from diagnostic errors. upload your labs, get a second analysis before making a life-changing decision every one of these has a customer already paying for a worse version of it. I compiled a detailed growth plan for each of these for myself drop which number you'd build first and I'll send it to you (maybe will discuss the terms how we can work together) Or even better DM me if you're not lazy and interesting guy to talk :<)
English
Andrej Karpathy just went on stage at Y Combinator and broke down why coding will never look the same again
he spent 39 minutes explaining what anyone with an idea and AI can build today - no engineering background needed
most people pay $1k - 2k for courses that don't even cover half of this
here's what he walks through:
> why "vibe coding" is replacing traditional development
> how one person with AI replaces an entire dev team
> the shift from writing syntax to giving clear instructions
> why the next generation of founders won't write a single line of code
the ones who get this early are already shipping products everyone else thinks needs a full team
that's why i wrote a complete 12-step playbook with 8 ready-to-use prompts to go from zero to a live app
the full guide is in the article below
KingWilliam@KingWilliamDefi
English
There's not a single country in the world that has avoided the throes of gentrification.
The rich hide in their concierge imbued ivory tower mansions ordering ubereats
While the poors doomscroll Short Form video and slowly get pushed out by immigrants
All according to keikaku.
朝は納豆@_asa710_
←足立区の運動会 →文京区の運動会
English
-> This girl set up AI helpers
-> to run her business with ai
-> By the time she wakes up
-> She has made $2000
-> She built the system once
-> It took a few evenings
-> Now it runs on its own
-> Here’s how it works.
-> One AI helper watches
-> what’s trending all day
-> It finds what people are
-> searching for, buying, and
-> asking about. It sends her
-> a report every morning, but
-> she barely reads it
-> anymore because the next
-> helper already acted on it.
-> Another helper writes
-> everything product lines
-> emails, social posts, video scripts
-> Hundreds of them.
-> All made to fit.
-> Posted before sunrise.
-> A third helper handles
-> customer messages.
-> It answers questions
-> handles requests, and
-> follows up. It’s quick, clear,
-> and never offline.
-> She wakes up, checks the
-> dashboard, and sees
-> what came in overnight.
-> Last month $23,000.
-> The month before $19,000
-> She’s not a coder.
-> She didn’t write a
-> single line of code.
-> She just explained what
-> she wanted in plain words,
-> and the helpers did the rest.
-> During the day she
-> makes small changes.
-> Adjusts things. Decides
-> where to point the system next.
-> That takes about an hour.
-> The other 23 hours, the
-> helpers keep working.
-> Most people trade time for money.
-> She traded one week of
-> setup for a business that never stops.
BeingInvested@0xbeinginvested
English
In 1 minute she shows how she makes $6,000 a month from kids videos
Everyone forgot about iPad kids.
But that is exactly where all the money is hiding right now.
These tiny channels post one video a day.
And they rack up absolutely insane amounts of views doing it.
For context, YouTube pays $200 to $3,000 per 100k views.
You open Picsart Flows.
Then you start a brand new video flow from scratch.
Type one simple prompt describing any cartoon animal you want.
Hit generate.
Out comes a whole batch of bright cartoon images in seconds.
Drop them into a clip where the animal hides, then pops out as a big reveal kids go crazy for.
Post one every single day.
That volume is exactly where the $6,000 a month comes from.
Save this before everyone floods YouTube with faceless kids channels.
winkle.@w1nklerr
English
Jack Ma says the next 10-15 years will destroy every business that hasn't changed
he built Alibaba from nothing to a $200B company
here's what you need to know:
> the problem isn't that manufacturing is bad - the problem is that outdated manufacturing is dead
> the future of business is not labor, it's data and AI
> companies that haven't adopted new technology in 10-15 years will be crying
> trade wars are wars for old manufacturing, not for the future
right now most companies think that because they went online - they're ready for the future
Jack Ma says that's the most dangerous illusion a business can have today
bookmark & watch today. 14 minutes. free.
Khairallah AL-Awady@eng_khairallah1
English
UKRAINIAN DEVELOPER BUILT AN OBSIDIAN BRAIN THAT READS HIS NOTES, TRADES POLYMARKET AND REPLIES TO HIS WIFE ON TELEGRAM WHILE HE SLEEPS
One vault, one CLAUDE.md and three systems running in parallel - Obsidian stores every thought, Hermes trades BTC 5-minute windows on Polymarket and a separate agent handles Telegram messages while he's in meetings.
The agent only enters a trade when the Markov persistence threshold hits p ≥ 0.87, Kelly criterion sizes the position automatically and win rate holds at 63-72% without a single emotional decision.
Every night Claude Opus reads the full trading journal, analyzes which states had the highest win rate and rewrites the strategy automatically - after 100 trades the agent is measurably smarter than it was at the start.
Three bots on the same math already made $2,112,019 in one market segment and his entire system costs under $10 a month.
His wife gets a reply on Telegram and doesn't know it's an agent.
0xRicker@0xRicker
English
A 23-year-old realtor from Nebraska lost his job and started studying Claude.
The result? He made $22,000 in a single day.
His Polymarket algorithm generated $249,000 in profit over the past month.
Dude spent around 250 hours teaching his Claude to collect data and run simulations through MiroFish.
Now the bot runs on full autopilot - the perfect machine for extracting mathematical inefficiencies in BTC markets.
He doesn’t need to predict the future - he simply calculates the perfect moment to buy both the UP and DOWN sides for a total of less than $1.
The crowd reacts to price direction quickly - but he’s even faster.
Thanks to data from private order books like Kraken and closed OTC BTC desks, he’s always one step ahead.
This isn’t "guessing where the chart will go"
This is engineered money. Pure fusion of AI + MiroFish + insane math on exclusive data.
Want to build your own algo? Save this post and read the article.
You only need Claude + laptop + 1 hour/day.
Giving This Free for 24 hours. To get it:
1. Comment Your thoughts .
2. Like and Retweet this post
3. Follow me @marryevan999
English
Claude turned $200 into $4,346 in 8 days.
No manual trades. Fully automated.
Most Polymarket traders think win rate matters.
It doesn’t.
Buying YES at 90¢ to make 10¢ looks smart until one loss wipes out multiple wins.
That’s why high win-rate traders often barely break even.
Claude ignores win rate.
Instead it tracks:
Realized PnL
Long-term consistency
Drawdown behavior
Category specialization
Risk-adjusted sizing
The filter is brutal:
• 90+ trades minimum
• 3+ months of steady growth
• Entries between 25–65¢
• Stable sizing
• Active in the last 4 days
Out of 800+ markets, only 57 wallets passed.
And the crazy part?
Some of the most profitable wallets had LOWER win rates.
Bot - t.me/KreoPolyBot?st…
Because they understood value.
They avoided overpriced 90¢ bets and waited for mispriced probabilities with real upside.
That’s the edge.
57 markets traded.
$200 → $4,346.
Not luck.
Just logic the market still hasn’t priced in.
Dropping the full workflow free for the next 24 hours.
Comment “CLAUDE” and I’ll send it over.
Anatoli Kopadze@AnatoliKopadze
English
In the agentic economy, once agents start moving value, I believe the most interesting question is: which infra the agents run on?
Rails decide everything here & I think it will condense to a core set of requirments:
- *Throughput* is the obvious one. An agent paying per API call, per query, per cycle of compute generates millions of micropayments an hour, which turns cost into a hard physical constraint. Sub cent payment means nothing if the fee to settle it is larger, so the real bar is enormous scale & a cost floor near zero at the same time.
- *Unified liquidity*, b/c no agent focused on optimizing the task at hand, is going to compound the problem he is dealing with by reasoning about which of 35 chains an asset lives on or holding a separate gas token for each, when an easy alternative exists. Fragmentation just produces failed or worse execution.
So what an agent will seek is a single liquidity surface it can express an intent against & have the routing handled beneath it, rather than a convoluted map of bridges it has to navigate itself.
- *Privacy* is crucial & perhaps still underrated. Consider that a public ledger broadcasts an agent's entire strategy to every competitor & mempool bot in real time.. He's asking to get front run & exploited.
So I'm thinking that in the agentic economy privacy becomes alpha preservation & is no longer just "ideological" or a "nice to have".
Also, if the agentic economy really takes off, privacy is a precondition for any enterprise/corporate issued agent, especially if they have to handle regulated data.
- *Identity*. Existing rails assume a human performing every txn, but an agent has no traditional ID, nothing a counterparty can verify. So the infrastructure itself will have to issue identity. Ideally some form of delegated authority with bounded & revocable spend & execution you can cryptographically prove that ran as intended. Otherwise, giving an agent a wallet is a liability rather than a capability. So it's a crucial piece of the puzzle.
So the job, as an investor bullish on the agentic economy, is now identifying the chain that best satisfies all these requirements:
T.U.L.I.P.
Throughput, Unified Liquidity, Identity, & Privacy
(might have picked a better acronym, but at least it's memorable)
English
Sam Altman, the CEO of OpenAI mentioned:
this is the best time in history for any 17-25 year old.
emphasizing how easy it is to build a business or startup with the tools available today.
in the article quoted below, I’ve documented 5 AI businesses that any 17-25 year old can venture into.
these business models are likely to remain relevant for years and could earn you as much as 6-7 figures yearly.
m0h@exploraX_
English
Shopify's CEO when he saw this leaked 4-hour course where a guy turned a brand new Shopify store into $50,000 in 30 days.
And realized anyone watching this can now do the same.
Which means a lot of new Shopify stores this week
Raytar@Raytar
He turned a new Shopify store into $50,000 in 30 days. Then recorded 4 hours showing the entire system. Every supplier, every angle, every tool. People sell this as a $497 course. He's giving it away for free below. Save this.
English
A girl set up AI agents to run her business while she sleeps.
By the time she wakes up, money has already moved.
She built the system once. Took a few evenings. Now it runs without her.
Here is what it looks like.
An AI agent monitors trending topics 24 hours a day.
Finds what people are searching for, what they are buying, what they need answered.
Sends her a report every morning she barely reads anymore because the next agent already acted on it.
Another agent writes the content. Product descriptions, emails, social posts, video scripts. Hundreds of them. Personalized. Published. Done before sunrise.
A third agent handles customer messages. Answers questions, processes requests, follows up. Fluent. Fast. Never offline.
She wakes up, checks the dashboard, sees what came in overnight.
Last month: $23,000.
The month before: $19,000.
She is not a programmer. She did not write a single line of code. She described what she wanted in plain language and the agents figured out the rest.
During the day she refines things. Adjusts. Thinks about where to point the system next.
That is maybe an hour of her time.
The other 23 hours the agents are working.
Most people trade time for money.
She traded one week of setup for a business that never clocks out.
Save this before everyone realizes agents do not need a salary.
Anatoli Kopadze@AnatoliKopadze
English
this guy built an ai-girl pipeline using real-time face filters, and d2c brands now pay him $2,000 per ugc video
he got tired of watching brands burn $4,000 on a single creator who takes 2 weeks to deliver one angle, so he built a setup that runs photoreal ai girls live from his own webcam, no actresses, no studios, no makeup artists
his monthly revenue hit $89,000 last month from a network of 7 ai personas across tiktok and instagram. the average ugc creator caps at $6k juggling 4 brand deals
the breakdown:
> hardware is the moat, but most people butcher the setup in the first frame. face mesh locked at 60fps with zero artifacting
> persona comes first, mess this up and nothing saves it: name, backstory, voice tone, niche before a single clip is shot
> face selection is not random. you a/b test features (eye spacing, jawline, hair contrast) because some faces convert better in 9:16
> you're picking who your audience trusts, not who looks cool. that's targeting baked into bone structure
> real-time physics run before the script, this is what kills the uncanny valley that destroys watch time in 2 seconds
> the filter has to survive the strap of a tank top, the texture of a knit cardigan, the hair flick
> batching is the move 96% skip: one performance, multiple personas, three platforms
> the system pushes 12 pieces before lunch while brands test 2 creators a week and wonder why their cpa sits at $94
the economics: each video costs $4 in compute, sells for $1,500 to $3,000, takes 14 minutes to produce. that's a 37,500% margin, while ugc agencies pay creators $400-800 per clip and net $200 after revisions
one supplement brand generated 14 variants with 7 personas in 4 hours and found a winner in 36 hours without flying a creator to la. they were paying $1,200 per ugc video and burning $6,000/week on content that didn't scale. now they spend $210 for 14 variants and their cpa dropped from $89 to $27
the avatars hold real products, warm window light on the persona, cold neon on the operator, mouth shapes sync to consonants not just vowels
just a webcam, a tracked face, and the discipline to move enough that the filter never has a chance to break
Kiyoro@0xKiyoro
English